{"cells":[{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["import datetime as dt\n","import time\n","\n","import IPython.display\n","import matplotlib.dates as mdates\n","import matplotlib.pyplot as plt\n","import matplotlib.ticker as ticker\n","import matplotlib_inline.backend_inline\n","import numpy as np\n","import pandas as pd\n","import requests as rq\n","import seaborn as sns\n","from ipywidgets import interact\n","\n","matplotlib_inline.backend_inline.set_matplotlib_formats('svg')\n","\n","def compose_url(**kwargs):\n","    baseurl = \"https://lapis.cov-spectrum.org/gisaid/v1/sample/aggregated?fields=date&host=Human&dateFrom=2021-11-01\"\n","    for key, value in kwargs.items():\n","        baseurl += \"&\" + key + \"=\" + value\n","    return baseurl\n","\n","\n","def get_data(**kwargs):\n","    url = compose_url(**kwargs)\n","    res = rq.get(url).json()\n","    return res\n","\n","def get_df(column_name, **kwargs):\n","    data = get_data(**kwargs)\n","    df = pd.DataFrame(data[\"data\"])\n","    df[\"date\"] = pd.to_datetime(df[\"date\"])\n","    df.set_index(\"date\", inplace=True)\n","    df.sort_index(inplace=True)\n","    df.rename(columns={\"count\": column_name}, inplace=True)\n","    return (df, data[\"info\"][\"dataVersion\"])\n","\n","\n","@interact(country=\"Denmark\", baseline=\"BA.*\", query=\"BA.2\")\n","def ratio_plot(country, baseline, query):\n","    print(\"Start update\")\n","    try:\n","        while True:\n","            base_df, base_version = get_df(\"baseline\", country=country, pangoLineage=baseline)\n","            query_df, query_version = get_df(\"query\", country=country, pangoLineage=query)\n","            if base_version == query_version:\n","                break\n","            else:\n","                print(\"Data version mismatch, retrying\")\n","                time.sleep(1)\n","        df = pd.concat([base_df, query_df], axis=1).fillna(0).astype(int)\n","        df[\"ratio\"] = df[\"query\"] / df[\"baseline\"]\n","        min = df[df[\"ratio\"] != 0].ratio.min()*0.3\n","        max = 1-(1-df[df[\"ratio\"] != 1].ratio.max())*0.7\n","        df[\"ratio\"] = df[\"ratio\"].clip(min,max)\n","        error = False\n","    except:\n","        error = True\n","        print(\"Error\")\n","    if error == False: \n","        print(\"Update plots\")\n","        IPython.display.clear_output()\n","        # print(df)\n","        fig, ax = plt.subplots(\n","            num=None, figsize=(8, 4), facecolor=\"w\", edgecolor=\"k\"\n","        )\n","        sns.scatterplot(\n","            data=df, x=\"date\", y=\"ratio\", hue=\"baseline\", size=\"baseline\"\n","        )\n","        fig.text(\n","            0.51,\n","            0.05,\n","            f\"Date: {str(dt.date.today())} | Data source: GISAID via covSpectrum | Viz: @CorneliusRoemer\",\n","            size=6,\n","            va=\"bottom\",\n","            ha=\"center\",\n","        )\n","        ax.set_xlim(left=df[df[\"query\"] != 0].index.min() - pd.Timedelta(days=1))\n","\n","        ax.set_yscale(\"logit\")\n","        # ax.set_ylim(\n","        #     bottom=min,\n","        #     top=max,\n","        # )\n","        ax.set_ylabel(f\"Proportion of {baseline} that are {query}\")\n","        ax.set_xlabel(\"Sample date\")\n","\n","        ax.set_title(f\"Share of {query} of all {baseline} in {country}\")\n","\n","        ax.get_legend().set_title(f\"{baseline} samples\")\n","        locator = mdates.AutoDateLocator()\n","        formatter = mdates.ConciseDateFormatter(locator)\n","        ax.xaxis.set_major_locator(locator)\n","        ax.xaxis.set_major_formatter(formatter)\n","        ax.xaxis.set_minor_locator(ticker.MultipleLocator(base=1.0))\n","        ax.grid(True, which=\"major\", linewidth=0.25)\n","        ax.grid(True, which=\"minor\", linewidth=0.1)\n","        ax.set_axisbelow(True)\n","\n","        percent_formatter = ticker.FuncFormatter(\n","            lambda y, _: f'{ np.format_float_positional(100*y, trim=\"-\", precision=6).rstrip(\".\")}%'\n","        )\n","        ax.yaxis.set_major_formatter(percent_formatter)\n","        ax.yaxis.set_minor_formatter(plt.NullFormatter())\n","        plt.subplots_adjust(left=0.15, right=0.9, top=0.9, bottom=0.20)\n","        plt.show()\n","        # plt.savefig(f\"{country}_{query}.png\", dpi=300)\n","\n"]}],"metadata":{"interpreter":{"hash":"82a7b143dffa2834bbebfae7ca04eb9403e273373f4b2947f57e402f018d2408"},"kernelspec":{"display_name":"Python 3.9.7 64-bit ('general': conda)","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.9.7"},"orig_nbformat":4},"nbformat":4,"nbformat_minor":2}